Background: Self-driving cars, also known as automated cars are a form of vehicle that can move without a driver or human involvement to control it. They employ numerous pieces of equipment to forecast the car’s navigation, and the car’s path is determined depending on the output of these devices. There are numerous methods available to anticipate the path of self-driving cars. Pedestrian detection is critical for autonomous cars to avoid fatalities and accidents caused by self-driving cars. Objectives: In this research, we focus on the algorithms in machine learning and deep learning to detect pedestrians on the roads. Also, to calculate the most accurate algorithm that can be used in pedestrian detection in automated cars by performing a literature review to select the algorithms. Methods: The methodologies we use are literature review and experimentation, literature review can help us to find efficient algorithms for pedestrian detection in terms of accuracy, computational complexity, etc. After performing the literature review we selected the most efficient algorithms for evaluation and comparison. The second methodology includes experimentation as it evaluates these algorithms under different conditions and scenarios. Through experimentation, we can monitor the different factors that affect the algorithms. Experimentation makes it possible for us to evaluate the algorithms using various metrics such as accuracy and loss which are mainly used to provide a quantitative measure of performance. Results: Based on the literature study, we focused on pedestrian detection deep learning models such as CNN, SSD, and RPN for this thesis project. After evaluating and comparing the algorithms using performance metrics, the outcomes of the experiments demonstrated that RPN was the highest and best-performing algorithm with 95.63% accuracy & loss of 0.0068 followed by SSD with 95.29% accuracy & loss of 0.0142 and CNN with 70.84% accuracy & loss of 0.0622. Conclusions: Among the three deep learning models evaluated for pedestrian identification, the CNN, RPN, and SSD, RPN is the most efficient model with the best performance based on the metrics assessed.

There has been a significant number of security concerns in recent times; as a result, security cameras have been installed to monitor activities and to prevent crimes in most public places. These analysis are done either through video analytic or forensic analysis operations on human observations. To this end, within the research context of this thesis, a proactive machine vision based military recognition system has been developed to help monitor activities in the military environment. The proposed object detection, recognition and re-identification systems have been presented in this thesis. A novel technique for military personnel recognition is presented in this thesis. Initially the detected camouflaged personnel are segmented using a grabcut segmentation algorithm. Since in general a camouflaged personnel's uniform appears to be similar both at the top and the bottom of the body, an image patch is initially extracted from the segmented foreground image and used as the region of interest. Subsequently the colour and texture features are extracted from each patch and used for classification. A second approach for personnel recognition is proposed through the recognition of the badge on the cap of a military person. A feature matching metric based on the extracted Speed Up Robust Features (SURF) from the badge on a personnel's cap enabled the recognition of the personnel's arm of service. A state-of-the-art technique for recognising vehicle types irrespective of their view angle is also presented in this thesis. Vehicles are initially detected and segmented using a Gaussian Mixture Model (GMM) based foreground/background segmentation algorithm. A Canny Edge Detection (CED) stage, followed by morphological operations are used as pre-processing stage to help enhance foreground vehicular object detection and segmentation. Subsequently, Region, Histogram Oriented Gradient (HOG) and Local Binary Pattern (LBP) features are extracted from the refined foreground vehicle object and used as features for vehicle type recognition. Two different datasets with variant views of front/rear and angle are used and combined for testing the proposed technique. For night-time video analytics and forensics, the thesis presents a novel approach to pedestrian detection and vehicle type recognition. A novel feature acquisition technique named, CENTROG, is proposed for pedestrian detection and vehicle type recognition in this thesis. Thermal images containing pedestrians and vehicular objects are used to analyse the performance of the proposed algorithms. The video is initially segmented using a GMM based foreground object segmentation algorithm. A CED based pre-processing step is used to enhance segmentation accuracy prior using Census Transforms for initial feature extraction. HOG features are then extracted from the Census transformed images and used for detection and recognition respectively of human and vehicular objects in thermal images. Finally, a novel technique for people re-identification is proposed in this thesis based on using low-level colour features and mid-level attributes. The low-level colour histogram bin values were normalised to 0 and 1. A publicly available dataset (VIPeR) and a self constructed dataset have been used in the experiments conducted with 7 clothing attributes and low-level colour histogram features. These 7 attributes are detected using features extracted from 5 different regions of a detected human object using an SVM classifier. The low-level colour features were extracted from the regions of a detected human object. These 5 regions are obtained by human object segmentation and subsequent body part sub-division. People are re-identified by computing the Euclidean distance between a probe and the gallery image sets. The experiments conducted using SVM classifier and Euclidean distance has proven that the proposed techniques attained all of the aforementioned goals. The colour and texture features proposed for camouflage military personnel recognition surpasses the state-of-the-art methods. Similarly, experiments prove that combining features performed best when recognising vehicles in different views subsequent to initial training based on multi-views. In the same vein, the proposed CENTROG technique performed better than the state-of-the-art CENTRIST technique for both pedestrian detection and vehicle type recognition at night-time using thermal images. Finally, we show that the proposed 7 mid-level attributes and the low-level features results in improved performance accuracy for people re-identification.

Dentre as diversas tecnologias de Assistência Avançada ao Condutor (ADAS) que têm sido adicionadas aos automóveis modernos estão os sistemas de detecção de pedestres. Tais sistemas utilizam sensores, como radares, lasers e câmeras de vídeo para captar informações do ambiente e evitar a colisão com pessoas no contexto do trânsito. Câmeras de vídeo têm se apresentado como um ótima opção para esses sistemas, devido ao relativo baixo custo e à riqueza de informações que capturam do ambiente. Muitas técnicas para detecção de pedestres baseadas em visão têm surgido nos últimos anos, tendo como característica a necessidade de um grande poder computacional para que se possa realizar o processamento das imagens em tempo real, de forma robusta, confiável e com baixa taxa de erros. Além disso, é necessário que sistemas que implementem essas técnicas tenham baixo consumo de energia, para que possam funcionar em um ambiente embarcado, como os automóveis. Uma tendência desses sistemas é o processamento de imagens de múltiplas câmeras presentes no veículo, de forma que o sistema consiga perceber potenciais perigos de colisão ao redor do veículo. Neste contexto, este trabalho aborda o coprojeto de hardware e software de uma arquitetura para detecção de pedestres, considerando a presença de quatro câmeras em um veículo (uma frontal, uma traseira e duas laterais). Com este propósito, utiliza-se a flexibilidade dos dispositivos FPGA para a exploração do espaço de projeto e a construção de uma arquitetura que forneça o desempenho necessário, o consumo de energia em níveis adequados e que também permita a adaptação a novos cenários e a evolução das técnicas de detecção de pedestres por meio da programabilidade. O desenvolvimento da arquitetura baseouse em dois algoritmos amplamente utilizados para detecção de pedestres, que são o Histogram of Oriented Gradients (HOG) e o Integral Channel Features (ICF). Ambos introduzem técnicas que servem como base para os algoritmos de detecção modernos. A arquitetura implementada permitiu a exploração de diferentes tipos de paralelismo das aplicações por meio do uso de múltiplos processadores softcore, bem como a aceleração de funções críticas por meio de implementações em hardware. Também foi demonstrada sua viabilidade no atendimento a um sistema contendo quatro câmeras de vídeo. / Among the several Advanced Driver Assistance (ADAS) technologies that have been added to modern vehicles are pedestrian detection systems. Those systems use sensors, such as radars, lasers, and video cameras to capture information from the environment and avoid collision with people in the context of traffic. Video cameras have become as a great option for such systems because of the relatively low cost and all of information they are able to capture from the environment. Many techniques for vison-based pedestrian detection have appeared in the last years, having as characteristic the necessity of a great computational power so that image can be processed in real time, in a robust and reliable way, and with low error rate. In addition, systems that implement these techniques require low power consumption, so they can operate in an embedded environment such as automobiles. A trend of these systems is the processing of images from multiple cameras mounted in vehicles, so that the system can detect potential collision hazards around the vehicle. In this context, this work addresses the hardware and software codesign of an architecture for pedestrian detection, considering the presence of four cameras in a vehicle (one in the front, one in the rear and two in the sides). For this purpose, the flexibility of FPGA devices is used for design space exploration and the construction of an architecture that provides the necessary performance, energy consumption at appropriate levels and also allows adaptation to new scenarios and evolution of pedestrian detection techniques through programmability. The development of the architecture was based on two algorithms widely used for pedestrian detection, which are Histogram of Oriented Gradients (HOG) and Integral Channel Features (ICF). Both introduce techniques that serve as the basis for modern detection algorithms. The implemented architecture allowed the exploration of different types of parallelism through the use of multiple softcore processors, as well as the acceleration of critical functions through implementations in hardware. It has also been demonstrated its feasibility in attending to a system containing four video cameras.

Feature extraction is the main driving force behind the advancement of the image processing techniques infields suchas image quality assessment, objectdetection, and object recognition. In this work, we perform a comprehensive and in-depth study on feature extraction for the following applications: image macro-uniformity assessment, 2.5D printing quality assessment, streak defect detection, and pedestrian detection. Firstly, a set of multi-scale wavelet-based features is proposed, and a quality predictor is trained to predict the perceived macro-uniformity. Secondly, the 2.5D printing quality is characterized by a set of merits that focus on the surface structure.Thirdly, a set of features is proposed to describe the streaks, based on which two detectors are developed: the first one uses Support Vector Machine (SVM) to train a binary classifier to detect the streak; the second one adopts Hidden Markov Model (HMM) to incorporates the row dependency information within a single streak. Finally, a novel set of pixel-difference features is proposed to develop a computationally efficient feature extraction method for pedestrian detection.

Dentre as diversas tecnologias de Assistência Avançada ao Condutor (ADAS) que têm sido adicionadas aos automóveis modernos estão os sistemas de detecção de pedestres. Tais sistemas utilizam sensores, como radares, lasers e câmeras de vídeo para captar informações do ambiente e evitar a colisão com pessoas no contexto do trânsito. Câmeras de vídeo têm se apresentado como um ótima opção para esses sistemas, devido ao relativo baixo custo e à riqueza de informações que capturam do ambiente. Muitas técnicas para detecção de pedestres baseadas em visão têm surgido nos últimos anos, tendo como característica a necessidade de um grande poder computacional para que se possa realizar o processamento das imagens em tempo real, de forma robusta, confiável e com baixa taxa de erros. Além disso, é necessário que sistemas que implementem essas técnicas tenham baixo consumo de energia, para que possam funcionar em um ambiente embarcado, como os automóveis. Uma tendência desses sistemas é o processamento de imagens de múltiplas câmeras presentes no veículo, de forma que o sistema consiga perceber potenciais perigos de colisão ao redor do veículo. Neste contexto, este trabalho aborda o coprojeto de hardware e software de uma arquitetura para detecção de pedestres, considerando a presença de quatro câmeras em um veículo (uma frontal, uma traseira e duas laterais). Com este propósito, utiliza-se a flexibilidade dos dispositivos FPGA para a exploração do espaço de projeto e a construção de uma arquitetura que forneça o desempenho necessário, o consumo de energia em níveis adequados e que também permita a adaptação a novos cenários e a evolução das técnicas de detecção de pedestres por meio da programabilidade. O desenvolvimento da arquitetura baseouse em dois algoritmos amplamente utilizados para detecção de pedestres, que são o Histogram of Oriented Gradients (HOG) e o Integral Channel Features (ICF). Ambos introduzem técnicas que servem como base para os algoritmos de detecção modernos. A arquitetura implementada permitiu a exploração de diferentes tipos de paralelismo das aplicações por meio do uso de múltiplos processadores softcore, bem como a aceleração de funções críticas por meio de implementações em hardware. Também foi demonstrada sua viabilidade no atendimento a um sistema contendo quatro câmeras de vídeo. / Among the several Advanced Driver Assistance (ADAS) technologies that have been added to modern vehicles are pedestrian detection systems. Those systems use sensors, such as radars, lasers, and video cameras to capture information from the environment and avoid collision with people in the context of traffic. Video cameras have become as a great option for such systems because of the relatively low cost and all of information they are able to capture from the environment. Many techniques for vison-based pedestrian detection have appeared in the last years, having as characteristic the necessity of a great computational power so that image can be processed in real time, in a robust and reliable way, and with low error rate. In addition, systems that implement these techniques require low power consumption, so they can operate in an embedded environment such as automobiles. A trend of these systems is the processing of images from multiple cameras mounted in vehicles, so that the system can detect potential collision hazards around the vehicle. In this context, this work addresses the hardware and software codesign of an architecture for pedestrian detection, considering the presence of four cameras in a vehicle (one in the front, one in the rear and two in the sides). For this purpose, the flexibility of FPGA devices is used for design space exploration and the construction of an architecture that provides the necessary performance, energy consumption at appropriate levels and also allows adaptation to new scenarios and evolution of pedestrian detection techniques through programmability. The development of the architecture was based on two algorithms widely used for pedestrian detection, which are Histogram of Oriented Gradients (HOG) and Integral Channel Features (ICF). Both introduce techniques that serve as the basis for modern detection algorithms. The implemented architecture allowed the exploration of different types of parallelism through the use of multiple softcore processors, as well as the acceleration of critical functions through implementations in hardware. It has also been demonstrated its feasibility in attending to a system containing four video cameras.

This master's thesis deals with pedestrian detection using neural networks in a video record captured by drone. Pedestrians are tracked, and their GPS coordinates are calculated using digital elevation models and mapped based on their identity and an information provided by the drone.

The topic of this thesis is the recognition and detection of moving object and persons in video sequence and in the static image. Designed application uses the combination of background model for movement detection, histograms of oriented gradients method for person recognition and Lucas-Kanade method for object tracking.

This thesis implements an automated system that counts pedestrians with 85% accuracy. Two approaches have been considered and evaluated in terms of count accuracy, cost and ease of deployment. The first approach employs the Autoscope Solo Terra, a traffic camera which is widely used to monitor vehicular traffic. The Solo Terra supports an image processing-based detector that counts the number of objects crossing user-defined areas in the captured image. The count is updated based on the amount of movement across the selected regions. Therefore, a second approach has been considered that uses a histogram of oriented gradients (HoG), an advanced vision based algorithm proposed by Dalal et al. which distinguishes a pedestrian from a non-pedestrian based on an omega shape formed by the head and shoulders of a human being. The implemented detection software processes video frames that are streamed from a low-cost digital camera. The frames are divided into sub-regions which are scanned for an omega shape whenever movement is detected in those regions. It has been found that the HoG-based approach degrades in performance due to occlusion under dense pedestrian traffic conditions whereas the Solo Terra approach appears to be more robust. Undercounts and overcounts were encountered using the Solo Terra approach. To combat the disadvantages of both the approaches, they were integrated to form a single system where count is incremented predominantly using the Solo Terra. The HoG-based approach corrects the obtained count under certain conditions. A preliminary prototype of the integrated system has been verified.

Die Datenbestände lokaler Fernsehsender umfassen oftmals mehrere zehntausend Videokassetten. Moderne Verfahren werden benötigt, um derartige Datenkollektionen inhaltlich automatisiert zu erschließen. Das Auffinden relevanter Objekte spielt dabei eine übergeordnete Rolle, wobei gesteigerte Anforderungen wie niedrige Fehler- und hohe Detektionsraten notwendig sind, um eine Korruption des Suchindex zu verhindern und erfolgreiche Recherchen zu ermöglichen. Zugleich müssen genügend Objekte indiziert werden, um Aussagen über den tatsächlichen Inhalt zu treffen. Diese Arbeit befasst sich mit der Anpassung und Optimierung bestehender Detektionsverfahren. Dazu wird ein auf die hohen Leistungsbedürfnisse der Videoanalyse zugeschnittenes holistisches Workflow- und Prozesssystem mit der Zielstellung implementiert, die Entwicklung von Bilderkennungsalgorithmen, die Visualisierung von Zwischenschritten sowie deren Evaluation zu ermöglichen. Im Fokus stehen Verfahren zur strukturellen Zerlegung von Videomaterialien und zur inhaltlichen Analyse im Bereich der Gesichtsdetektion und Fußgängererkennung. / The data collections of local television stations often consist of multiples of ten thousand video tapes. Modern methods are needed to exploit the content of such archives. While the retrieval of objects plays a fundamental role, essential requirements incorporate low false and high detection rates in order to prevent the corruption of the search index. However, a sufficient number of objects need to be found to make assumptions about the content explored. This work focuses on the adjustment and optimization of existing detection techniques. Therefor, the author develops a holistic framework that directly reflects on the high demands of video analysis with the aim to facilitate the development of image processing algorithms, the visualization of intermediate results, and their evaluation and optimization. The effectiveness of the system is demonstrated on the structural decomposition of video footage and on content-based detection of faces and pedestrians.

Cette thèse s'inscrit dans le contexte de la vision embarquée pour la détection et la reconnaissance d'obstacles routiers, en vue d'application d'assistance à la conduite automobile.A l'issue d'une étude bibliographique, nous avons constaté que la problématique de détection d'obstacles routiers, notamment des piétons, à l'aide d'une caméra embarquée, ne peut être résolue convenablement sans recourir aux techniques de reconnaissance de catégories d'objets dans les images. Ainsi, une étude complète du processus de la reconnaissance est réalisée, couvrant les techniques de représentation,de classification et de fusion d'informations. Les contributions de cette thèse se déclinent principalement autour de ces trois axes.Notre première contribution concerne la conception d'un modèle d'apparence locale basée sur un ensemble de descripteurs locaux SURF (Speeded Up RobustFeatures) représentés dans un Vocabulaire Visuel Hiérarchique. Bien que ce modèle soit robuste aux larges variations d'apparences et de formes intra-classe, il nécessite d'être couplé à une technique de classification permettant de discriminer et de catégoriser précisément les objets routiers. Une deuxième contribution présentée dans la thèse porte sur la combinaison du Vocabulaire Visuel Hiérarchique avec un classifieur SVM.Notre troisième contribution concerne l'étude de l'apport d'un module de fusion multimodale permettant d'envisager la combinaison des images visibles et infrarouges.Cette étude met en évidence de façon expérimentale la complémentarité des caractéristiques locales et globales ainsi que la modalité visible et celle infrarouge.Pour réduire la complexité du système, une stratégie de classification à deux niveaux de décision a été proposée. Cette stratégie est basée sur la théorie des fonctions de croyance et permet d'accélérer grandement le temps de prise de décision.Une dernière contribution est une synthèse des précédentes : nous mettons à profit les résultats d'expérimentations et nous intégrons les éléments développés dans un système de détection et de suivi de piétons en infrarouge-lointain. Ce système a été validé sur différentes bases d'images et séquences routières en milieu urbain. / The aim of this thesis arises in the context of Embedded-vision system for road obstacles detection and recognition : application to driver assistance systems. Following a literature review, we found that the problem of road obstacle detection, especially pedestrians, by using an on-board camera, cannot be adequately resolved without resorting to object recognition techniques. Thus, a preliminary study of the recognition process is presented, including the techniques of image representation, Classification and information fusion. The contributions of this thesis are organized around these three axes. Our first contribution is the design of a local appearance model based on SURF (Speeded Up Robust Features) features and represented in a hierarchical Codebook. This model shows considerable robustness with respect to significant intra-class variation of object appearance and shape. However, the price for this robustness typically is that it tends to produce a significant number of false positives. This proves the need for integration of discriminative techniques in order to accurately categorize road objects. A second contribution presented in this thesis focuses on the combination of the Hierarchical Codebook with an SVM classifier.Our third contribution concerns the study of the implementation of a multimodal fusion module that combines information from visible and infrared spectrum. This study highlights and verifies experimentally the complementarities between the proposed local and global features, on the one hand, and visible and infrared spectrum on the other hand. In order to reduce the complexity of the overall system, a two-level classification strategy is proposed. This strategy, based on belieffunctions, enables to speed up the classification process without compromising there cognition performance. A final contribution provides a synthesis across the previous ones and involves the implementation of a fast pedestrian detection systemusing a far-infrared camera. This system was validated with different urban road scenes that are recorded from an onboard camera.